The invention relates to a novel methods for measuring cellular ion channel transmission and methods and compositions useful in the identification of ligand gated ion channel agonists and modulators.
Ion channel proteins form hydrophilic pores that extend across the cellular lipid bilayer; when these pores are open, they allow specific molecules (usually inorganic ions of appropriate size and charge) to pass through them and thereby cross the membrane.
Channel proteins which are concerned specifically with inorganic ion transport are referred to as ion channels, and include ion channels for sodium, potassium, calcium, and chloride ions. Ion channels which open in response to a change in the voltage across the membrane are referred to as voltage gated ion channels (or voltage-dependent ion channels). Ion channels which open in response to the binding of a ligand to the channel protein are referred to as ligand gated ion channels.
The present invention describes new ion channels and provides methods and compositions suitable for high throughput screening of ion channels.
Voltage Gated Sodium Channel
Voltage gated ion channels are a class of channel proteins that play a major role in cellular electrical excitability. In the majority of excitable tissues, the early depolarization phase of action potentials is mediated by a sodium current via voltage-dependent sodium channels (also known as voltage-gated sodium channels or VGSCs). The sodium channel is one of the most thoroughly characterized of the voltage gated channels. The primary structures of many sodium channels from a variety of tissues (brain, skeletal muscle and cardiac muscle) and organisms (jellyfish, squid, eel, rat, human) have been identified, and their amino acid sequences show individual regions which are highly conserved over evolution, indicating that voltage-dependent sodium channels belong to a large superfamily of evolutionarily related proteins. All published polypeptide complexes of VGSCs have in common a large, about 260 kDa glycoprotein (the pore forming subunit) which is called the alpha subunit (Agnew et al. 1978; Agnew et al. 1980; Catterall 1986; Catterall 1992). Additional lower molecular weight polypeptides, the beta-subunits, have been found to be associated with sodium channels from mammalian muscle (Kraner et al. 1985; Tanaka et al. 1983) and brain (Hartshorne and Catterall 1984). The large, pore-forming alpha subunit is sufficient for all known functions of VGSCs (Catterall 1992) while the beta subunits modulate some of the functions of the alpha subunit (Catterall 1992).
Voltage Gated Potassium Channels
Voltage-gated potassium channels make up a large molecular family of integral membrane proteins that are fundamentally involved in the generation of bioelectric signals such as nerve impulses. These proteins span the cell membrane, forming potassium-selective pores that are rapidly switched open or closed by changes in membrane voltage. Several chemical entities have been discovered to be potent and specific openers of vascular potassium K+ channels. These include cromakalim and its derivatives and RP 52891. This mechanism is also shared, at least partially, by drugs such as minoxidil, diazoxide, pinacidil and nicorandil. The opening of plasmalemmal K+ channels produces loss of cytosolic K+. This effect results in cellular hyperpolarization and functional vasorelaxation. In normotensive or hypertensive rats, K+ channel activators decrease aortic blood pressure (by producing a directly mediated fall in systemic vascular resistance) and reflexively increase heart rate. K+ channel openers produce selective coronary vasodilatation and afford functional and biochemical protection to the ischemic myocardium.
The structure of a typical voltage-gated potassium channel protein is known to be comprised of six membrane spanning domains in each subunit, each of which is regulated by changes in membrane potential. B. Hille. xe2x80x9cIonic Channels of Excitable Membranesxe2x80x9d(Sinauer, Sunderland, Mass., 1992). Voltage-gated potassium channels sense changes in membrane potential and move potassium ions in response to this alteration in the cell membrane potential. Molecular cloning studies on potassium channel proteins has yielded information primarily for members of the voltage-gated family of potassium channels. Various genes encoding these voltage-gated family of potassium channel proteins have been cloned using Drosophila genes derived from both the Shaker, Shaw and Shab loci; Wei, A. et. al., Science (1990) Vol. 248 pp. 599-603.
Voltage Gated Calcium Channels
Voltage-gated calcium channels are present in neurons, and in cardiac, smooth, and skeletal muscle and other excitable cells. These channels are known to be involved in membrane excitability, muscle contraction, and cellular secretion, such as in exocytotic synaptic transmission (McCleskey, et al.,1987). In neuronal cells, voltage-gated calcium channels have been classified by their electrophysiological as well as by their biochemical (binding) properties.
Calcium channels are generally classified according to their electrophysiological properties as Low-voltage-activated (LVA) or High-voltage-activated (HVA) channels. HVA channels are currently known to comprise at least three groups of channels, known as L-, N- and P-type channels (Nowycky, et al., 1985). These channels have been distinguished one from another structurally and electrophysiologically as well as biochemically on the basis of their pharmacology and ligand binding properties. Thus, dihydropyridines, diphenylalkylamines and piperidines bind to the alpha1 subunit of the L-type calcium channel and block a proportion of HVA calcium currents in neuronal tissue, which are termed L-type calcium currents.
N- or omega-type HVA calcium channels are distinguishable from other calcium channels by their sensitivity to omega conotoxins (omega conopeptides). Such channels are insensitive to dihydropyridine compounds, such as L-type calcium channel blockers nimodipine and nifedipine. (Sher and Clementi, 1991).
Ligand Gated Ion Channel Receptors
Ligand-gated ion channels provide a means for communication between cells of the central nervous system. These channels convert a signal (e.g., a chemical referred to as a neurotransmitter) that is released by one cell into an electrical signal that propagates along a target cell membrane. A variety of neurotransmitters and neurotransmitter receptors exist in the central and peripheral nervous systems. At the present time, numerous families of ligand-gated receptors have been identified and characterized on the basis of sequence identity these include nicotinic acetylcholine, glutamate, glycine, GABA A, 5-HT3, and the purinoceptors. These can be further characterized by whether the gated ion channel transmits cations or anions. Those which form cationic channels include, for example, excitatory nicotinic acetylcholine receptors (nAChRs), excitatory glutamate-activated receptors, the 5-HT3 serotonin receptor, and the purine receptor.
Those which form anionic channels include, for example, the inhibitory GABA and glycine-activated receptors. This discussion will confine itself to those ligand gated ion channel receptors which conduct cations.
5HT3 Receptor
Molecular cloning has indicated that serotonin (5-hydroxytryptamine, also referred to as 5-HT) receptors belong to at least two protein superfamilies: G-protein-associated receptors and ligand-gated ion channel. The 5-HT3 receptor belongs to the family of ligand-gated ion channels. As discussed below the 5-HT3 receptor is primarily a sodium potassium ligand gated ion channel under physiologic conditions. The inflammatory and painproducing effects of serotonin are generally believed to be mediated via 5HT3 receptors on peripheral sensory endings (Richardson, B. P., et al., 1985).
Nicotinic Receptors
The nicotinic acetylcholine receptors (nAChRs) are multisubunit proteins of neuromuscular and neuronal origins. These receptors form ligand-gated ion channels that mediate synaptic transmission between nerve and muscle and between neurons upon interaction with the neurotransmitter acetylcholine (ACh). Since various nicotinic acetylcholine receptor (nAChR) subunits exist, a variety of nAChR compositions (i.e., combinations of subunits) exist. The different nAChR compositions exhibit different specificities for various ligands and are thereby pharmacologically distinguishable. Thus, the nicotinic acetylcholine receptors expressed at the vertebrate neuromuscular junction in vertebrate sympathetic ganglia and in the vertebrate central nervous system have been distinguished on the basis of the effects of various ligands that bind to different nAChR compositions. For example, the elapid alpha-neurotoxins that block activation of nicotinic acetylcholine receptors at the neuromuscular junction do not block activation of some neuronal nicotinic acetylcholine receptors that are expressed on several different neuron-derived cell lines.
Muscle nAChR is a glycoprotein composed of five subunits with the stoichiometry alpha 2 alpha (gamma or epsilon) delta. Each of the subunits has a mass of about 50-60 kilodaltons (kd) and is encoded by a different gene. The alpha 2 beta (gamma or epsilon) delta complex forms functional receptors containing two ligand binding sites and a ligand-gated transmembrane channel. Upon interaction with a cholinergic agonist, muscle nicotinic AChRs conduct sodium ions. The influx of sodium ions rapidly short-circuits the normal ionic gradient maintained across the plasma membrane, thereby depolarizing the membrane. By reducing the potential difference across the membrane, a chemical signal is transduced into an electrical signal that signals muscle contraction at the neuromuscular junction.
Functional muscle nicotinic acetylcholine receptors have been formed with alpha beta delta gamma subunits, alpha beta gamma subunits, alpha beta delta subunits, alpha beta gamma subunits or alpha delta subunits, but not with only one subunit (see e.g., Kurosaki et al. 1987; Camacho et al. 1993) In contrast, functional neuronal AChRs (nAChRs) can be formed from alpha subunits alone or combinations of alpha and beta subunits. The larger alpha subunit is generally believed to be the ACh-binding subunit and the lower molecular weight beta subunit is generally believed to be the structural subunit, although it has not been definitively demonstrated that the beta subunit does not have the ability to bind ACh. Each of the subunits which participate in the formation of a functional ion channel are, to the extent they contribute to the structure of the resulting channel, xe2x80x9cstructuralxe2x80x9d subunits, regardless of their ability (or inability) to bind ACh.
Neuronal AChRs (nAChRs), which are also ligand-gated ion channels, are expressed in ganglia of the autonomic nervous system and in the central nervous system (where they mediate signal transmission), in post-synaptic locations (where they modulate transmission), and in pre- and extra-synaptic locations (where they may have additional functions). The nAChRs comprise a large family of neurotransmitter regulated ion channels that control neuronal activity and brain function. These receptors have a pentameric structure. The gene family is composed of nine alpha and four beta subunits that co-assemble to form multiple subtypes of receptors that have a distinctive pharmacology. Acetycholine is the endogenous regulator of all of the subtypes, while nicotine non-selectively activates all nAChRs. Known chemical templates have subtype selectivity.
xcex17 nAChR is a ligand-gated Ca++ channel formed by a homopentamer of xcex17 subunits. xcex17 nAChR is of particular interest because xcex17 nAChR agonists increase neurotransmitter release, increase cognition, arousal, attention, learning and memory. xcex17 nAChR is expressed at high levels in the hippocampus, ventral tegmental area and ascending cholinergic projections from nucleus basilis to thalamocortical areas. Previous studies have established that a xcex1-bungarotoxin (xcex1-btx) binds selectively to this homopetameric, xcex17 nAChR subtype, and that xcex17 nAChR has a high affinity binding site for both xcex1-btx and methyllycaconitine (MLA). We have chosen to use xcex17 nAChR as a model system for high throughput drug screening
Glutamate Receptors
Glycine also functions in excitatory transmission by modulating the actions of glutamate, the major excitatory neurotransmitter in the central nervous system. (Johnson and Ascher, 1987)
Glutamate binds or interacts with one or more glutamate receptors which can be differentiated pharmacologically into several subtypes. In the mammalian central nervous system (CNS) there are three main subtypes of ionotropic glutamate receptors, defined pharmacologically by the selective agonists N-methyl-D-aspartate (NMDA), kainate (KA), and alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA). The NMDA receptor has been implicated in a variety of neurological pathologies including stroke, head trauma, spinal cord injury, epilepsy, anxiety, and neurodegenerative diseases such as Alzheimer""s Disease (Watkins and Collingridge 1989). A role for NMDA receptors in nociception and analgesia has been postulated as well (Dickenson, 1990). More recently, AMPA receptors have been widely studied for their possible contributions to such neurological pathologies (Fisher and Bogousslavsky, 1993).
When activated by glutamate, the endogenous neurotransmitter, the NMDA receptor permits the influx of extracellular calcium (Ca++) and sodium (Na+) through an associated ion channel. The NMDA receptor allows considerably more influx of Ca++ than do kainate or AMPA receptors and is an example of a receptor-operated Ca++ channel. Normally, the channel is opened only briefly, allowing a localized and transient increase in the concentration of intracellular Calcium (Ca++) which, in turn, alters the functional activity of the cell.
The activity of the NMDA receptor-ionophore complex is regulated by a variety of modulatory sites that can be targeted by selective antagonists. Competitive antagonists, such as the phosphonate AP5, act at the glutamate binding site, whereas noncompetitive antagonists, such as phencyclidine (PCP), MK-801 or magnesium (Mg++), act within the associated ion channel (ionophore). There is also a glycine binding site that can be blocked selectively with compounds such as 7-chlorokynurenic acid. There is evidence suggesting that glycine acts as a co-agonist, so that both glutamate and glycine are necessary to fully elicit NMDA receptor-mediated responses. Other potential sites for modulation of NMDA receptor function include a zinc (Zn less than 2+ greater than ) binding site and a sigma ligand binding site. Additionally, endogenous polyamines such as spermine are believed to bind to a specific site and so potentiate NMDA receptor function (Ransom and Stec, 1988). The potentiating effect of polyamines on NMDA receptor function may be mediated via a specific receptor site for polyamines.
Purinergic Receptors
Purinergic receptors are classified as P1 (adenosine as ligand) and P2 (ATP as ligand). The P2 receptors are subclassified into two broad types-those that are 7-transmembrane receptors that couple to G-proteins (P 2Y, P 2U, P 2T, and perhaps P 2Z. Another major class of purinoceptors are the P2x purinoceptors which are ligand-gated ion channels possessing intrinsic ion channels permeable to Na+, K+, and Ca++. P2x receptors described in sensory neurons are important for primary afferent neurotransmission and nociception. ATP is known to depolarize sensory neurons and plays a role in nociceptor activation since ATP released from damaged cells stimulates P2x receptors leading to depolarization of nociceptive nerve-fiber terminals. ATP-sensitive potassium channels have been discovered in numerous tissues, including kidney, vascular and non-vascular smooth muscle and brain, and binding studies using radiolabeled ligands have confirmed their existence. Opening of these channels causes potassium (K less than + greater than ) efflux and hyperpolarizes the cell membrane
Ion Channels as Drug Targets
Ion channels both ligand gated and voltage gated, are in general excellent and validated drug targets. For some channels however, a functional high throughput screening assay is problematic because expression levels are low and function is hard to measure using standard detection technology for high throughput screening. For those channels which normally conduct a cation other than calcium high througput screening methods are often cumbersome. For calcium conductance however, several rapid assays exist. It would often be desireable to This invention provides the scientist with a detailed description about how to convert a channel normally conducting sodium or potassium under physiologic conditions to one conducting calcium for ease in assay development.
The xcex17 nAChR discussed above is one ligand gated ion channel that has proved to be a difficult target for developing a functional high throughput screening assay. Native xcex17 nAChR are not routinely able to be stably expressed in most mammalian cell lines (Cooper and Millar 1997). Repeated attempts by our group to stably express the human xcex17 nAChR in HRK 293, CHO, COS and SH-EP1 were unsuccessful. While it was possible to identify cell lines that initially expressed functional xcex17 nAChR, these lines dramatically lost receptor expression with prolonged growth in culture. Under these conditions it was not possible to use these lines for screening purposes. Another feature that makes functional assays of xcex17 nAChR challenging is that the receptor is rapidly (100 milliseconds) inactivated agonist application. This rapid inactivation greatly limits the functional assays that can be used to measure channel activity
One solution to the problem is to engineer the xcex17 nAChR to have a longer duration of open probability and to have it be expressed better in mammalian cells. We are aware of a report indicating that a chimeric receptor formed between the N-terminal ligand binding domain of the xcex17 nAChR (AA 1-201) and the pore forming C-terminal domain of the 5-HT3 receptor expressed well in Xenopus oocytes while retaining nicotinic agonist sensitivity (Eisele et al. 1993). Eisele et al (1993) used the N-terminus of the avian (chick) form of the xcex17 nAChR receptor and the c-terminus of the mouse form of the 5-HT3 gene. The report of Eisele et. al. was interesting to us because we knew from our own studies that the 5-HT3 channels expressed well in most mammalian cells. In addition, we also knew from past studies that 5-HT3 channels inactivated much slower than nicotinic channels. A chimeric receptor prepared from the ligand binding region of xcex17 nAChR and the pore forming domain of 5-HT3 might express well in mammalian cells and might be easier to measure in a functional assay. However, under physiological conditions the xcex17 nAChR is a calcium channel while the 5-HT3 receptor is a sodium and potassium channel. Indeed, Eisele et al. teaches that the chicken xcex17 nAChR/ mouse 5-HT3 receptor behaves quite differently than the native xcex17 nAChR with the pore element not conducting calcium but actually being blocked by calcium ions. The chicken/mouse hybrid of Eisele is also not suitable for accessing compounds for their activity at the human xcex17 nAChR receptor. The human xcex17 nAChR has 92% identity with the chicken xcex17 nAChR, but surprisingly, the pharmacology of the two receptors are different. For example, 1,1-dimethyl-4-phenylpiperazinium is a full agonist at the human receptor and a partial agonist at the chicken receptor (Peng et al 1994). Other large species-specific differences in binding affinity have been noted (Peng et al 1994).
Ligand binding can be accessed in either whole cells or membrane preparations but both kinds of assays are cumbersome. Whole cell assays have been difficult to perform in a high throughput screening format because of the extensive washing and manipulation required to obtain a good signal to noise ratio. Isolated membranes have been used in such assays but also typically require extensive manipulation to prepare the membranes themselves and the assay itself requires extensive manipulation and washing to obtain a favorable signal to noise ratio. Such assays are illustrated in U.S. Pat. No. 6,022,704. A binding assay which could be performed without such required extensive manipulation would be extremely useful.
Within the last few years very precise measurement of cellular fluorescence in a high throughput whole cell assay has become possible with the use of a device marketed by Molecular Devices, Inc. designated xe2x80x9cFLIPRxe2x80x9d (Schroeder et al. 1996), entire document, full reference provided below, incorporated herein by reference. FLIPR has shown considerable utility in measuring membrane potential of mammalian cells using voltage-sensitive fluorescent dyes but is useful for measuring essentially any cellular fluorescence phenomenon. The device uses low angle laser scanning illumination and a mask to selectively excite fluorescence within approximately 200 microns of the bottoms of the wells in standard 96 well plates. The low angle of the laser reduces background by selectively directing the light to the cell monolayer. This avoids background fluorescence of the surrounding media. This system then uses a CCD camera to image the whole area of the plate bottom to measure the resulting fluorescence at the bottom of each well. The signal measured is averaged over the area of the well and thus measures the average response of a population of cells. The system has the advantage of measuring the fluorescence in each well simultaneously thus avoiding the imprecision of sequential measurement well by well measurement. The system is also designed to read the fluorescent signal from each well of a 96 or 384 well plate as fast as twice a second. This feature provides FLIPR with the capability of making very fast measurements in parallel. This property allows for the measurement of changes in many physiological properties of cells that can be used as surrogated markers to a set of functional assays for drug discovery. FLIPR is also designed to have state of the art sensitivity. This allows it to measure very small changes with great precision.
U.S. Pat. No. 6,022,704, Feb. 8, 2000, DNA and mRNA encoding an alpha 4 subunit of human neuronal nicotinic acetylcholine receptor and cells transformed with same, Elliott, K. J. et. al.
1. Agnew, W. S., et al., Purification of the tetrodotoxin-binding component associated with the voltage-sensitive sodium channel from Electrophorus electricus electroplax membranes. Proc Natl Acad Sci U S A, 1978. 75(6): p. 2606-10.
2. Agnew, W. S., et al., Identification of a large molecular weight peptide associated with a tetrodotoxin binding protein from the electroplax of Electrophorus electricus. Biochem Biophys Res Commun, 1980. 92(3): p. 860-6.
3. Brown, A. M., et al., Ion permeation and conduction in a human recombinant 5-HT3 receptor subunit (h5-HT3A). J Physiol (Lond), 1998. 507(Pt 3): p. 653-65.
4. Camacho, P., et al., The epsilon subunit confers fast channel gating on multiple classes of acetylcholine receptors. J Neurosci, 1993. 13(2): p. 605-13.
5. Catterall, W. A., Molecular properties of voltage-sensitive sodium channels. Annu Rev Biochem, 1986. 55: p. 953-85.
6. Catterall, W. A., Cellular and molecular biology of voltage-gated sodium channels. Physiol Rev, 1992. 72(4 Suppl): p. S15-48.
7. Catterall, W. and P. N. Epstein, Ion channels. Diabetologia, 1992. 35 Suppl 2: p. S23-33.
8. Cooper, S. T. and N. S. Millar, Host cell-specific folding and assembly of the neuronal nicotinic acetylcholine receptor alpha7 subunit. J Neurochem, 1997. 68(5): p. 2140-51.
9. Dickenson, A. H., A cure for wind up: NMDA receptor antagonists as potential analgesics. Trends Pharmacol Sci, 1990. 11(8): p. 307-9.
10. Eisele, J. L., et al., Chimaeric nicotinic-serotonergic receptor combines distinct ligand binding and channel specificities [see comments]. Nature, 1993. 366(6454): p. 479-83.
11. Fisher, M. and J. Bogousslavsky, Evolving toward effective therapy for acute ischemic stroke. Jama, 1993. 270(3): p. 360-4.
12. Hartshorne, R. P. and W. A. Catterall, The sodium channel from rat brain. Purification and subunit composition. J Biol Chem, 1984. 259(3): p. 1667-75.
13. Hille, B., Ionic Channels of Excitable Membranes. 1992.
14. Holladay, M. W., et al., Identification and initial structure-activity relationships of (R)-5-(2-azetidinylmethoxy)-2-chloropyridine (ABT-594), a potent, orally active, non-opiate analgesic agent acting via neuronal nicotinic acetylcholine receptors. J Med Chem, 1998. 41(4): p. 407-12.
15. Holladay, M. W., et al., Structure-activity studies related to ABT-594, a potent nonopioid analgesic agent: effect of pyridine and azetidine ring substitutions on nicotinic acetylcholine receptor binding affinity and analgesic activity in mice. Bioorg Med Chem Lett, 1998. 8(19): p. 2797-802.
16. Johnson, J. W. and P. Ascher, Glycine potentiates the NMDA response in cultured mouse brain neurons. Nature, 1987. 325(6104): p. 529-31.
17. Kraner, S. D., J. C. Tanaka, and R. L. Barchi, Purification and functional reconstitution of the voltage-sensitive sodium channel from rabbit T-tubular membranes. J Biol Chem, 1985. 260(10): p. 6341-7.
18. Kurosaki, T., et al., Functional properties of nicotinic acetylcholine receptor subunits expressed in various combinations. FEBS Lett, 1987. 214(2): p. 253-8.
19. McCleskey, E. W., et al., Omega-conotoxin: direct and persistent blockade of specific types of calcium channels in neurons but not muscle. Proc Natl Acad Sci U S A, 1987. 84(12): p. 4327-31.
20. Nowycky, M. C., A. P. Fox, and R. W. Tsien, Three types of neuronal calcium channel with different calcium agonist sensitivity. Nature, 1985. 316(6027): p. 440-3.
21. Peng, X., et al., Human alpha 7 acetylcholine receptor: cloning of the alpha 7 subunit from the SH-SY5Y cell line and determination of pharmacological properties of native receptors and functional alpha 7 homomers expressed in Xenopus oocytes. Mol Pharmacol, 1994. 45(3): p. 546-54.
22. Ransom, R. W. and N. L. Stec, Cooperative modulation of [3H]MK-801 binding to the N-methyl-D-aspartate receptor-ion channel complex by L-glutamate, glycine, and polyamines. J Neurochem, 1988. 51(3): p. 830-6.
23. Richardson, B. P., et al., Identification of serotonin M-receptor subtypes and their specific blockade by a new class of drugs. Nature, 1985. 316(6024): p. 126-31.
24. Sher, E. and F. Clementi, Omega-conotoxin-sensitive voltage-operated calcium channels in vertebrate cells. Neuroscience, 1991. 42(2): p. 301-7.
25. Tanaka, J. C., J. F. Eccleston, and R. L. Barchi, Cation selectivity characteristics of the reconstituted voltage-dependent sodium channel purified from rat skeletal muscle sarcolemma. J Biol Chem, 1983. 258(12): p. 7519-26.
26. Watkins, J. C. and G. L. Collingridge, The NMDA Receptor. First ed. 1989: IRL Press.
27. Wei, A., et al., K+ current diversity is produced by an extended gene family conserved in Drosophila and mouse. Science, 1990. 248(4955): p. 599-603.
28. Yamaguchi, S., S. D. Donevan, and M. A. Rogawski, Anticonvulsant activity of AMPA/kainate antagonists: comparison of GYKI 52466 and NBOX in maximal electroshock and chemoconvulsant seizure models. Epilepsy Res, 1993. 15(3): p. 179-84.
29. Yang, J., Ion permeation through 5-hydroxytryptamine-gated channels in neuroblastoma N18 cells. J Gen Physiol, 1990. 96(6): p. 1177-98.
30. Schroeder et al., FLIPR: A New Instrument for Accurate, High Throughput Optical Screening, Journal of Biomolecular Screening, 1996, 1(2) pp.75-80 (incorporated herein by reference).